Consider an example in which we measure the percent body fat, height (in inches), and waist size (in inches) of a group of people. What might you expect the simple regression relating percent body fat to height would look like? An example for one sample might be:

Body Fat % = 13.012 − 0.031 × Height
Line Fit Plot
A graph of height, from 55 to 80, to percent body fat, from 0 to 25, plots points with no correlation to the nearly horizontal predicted line.

In this example, there is almost no discernible relationship between height and percent body fat. This information could be useful, but suppose we still want to come up with a good model for predicting body fat.